Week 1

Week 2

Week 4

Week 5

Week 6

Week 7

Week 8

Week 9

Week 10

Week 3 (June 3 - 9)

I spent this week putting together another CPHS packet which updated our interaction with dual households.  Jen and I realized we need to receive food consumption questionnaires from both members of dual households in order to accurately analyze the reported food consumption vs food purchases.  Dual households also can include families of one parent and one child, so we had to update the CPHS paperwork in regard to children.  Doing this submission took a few days since I had to no references for protocols involving children.  I had to clarify that children would not be interviewed or observed, and that we would not pay dual households more for their two questionnaires. 

I also had to contact our pilot dual household and ask if the second member of the household would be willing to complete the questionnaire.  It took a few days for the household to respond, and the second member ultimately decided not to complete the questionnaire.  It was our bad luck, but we went on with the pilot.

This week I also investigated database alternatives to our MySQL one.  Because our food database is so large, the MySQL queries could be slow.  I investigated the possibility of using Oracle or Microsoft Access.  Oracle is too large-scale for us.  Access is a good possibility; speed would be less of an issue.  Because it is Microsoft though, we would have less control over the code.  It is helpful that the USDA has their data already in Access, and I downloaded the database without any problems.  I discussed these options with Jen.  We decided not to make any changes at the present moment. 

I also created for my own reference a schematic of our current database.  This took a few days since the database is pretty extensive, with many tables, and the relationship between tables/primary keys was not always clear to me.  I have a better idea now.

I entered the candidates data into an excel spreadsheet. Excel has the option of saving as CSV (comma delimited), and this will be necessary when we export he data to a stat program, such as SPSS. I also calculated the Body Mass Index (BMI) for each candidate. BMI = (weight in kg)/(height in meters)squared. This information could be important in selecting long-term survey participants

This week I also read Jakob Nielsen's paper on Heuristic Evaluation and his Summary of Usability Inspection Methods, to start thinking about the overall interface design of the device.  

< Previous Week          Next Week >